BLOG POST

Indigenous Peoples Prevent Deforestation. What About Other Local Communities?

August 12, 2014

Just how good are Indigenous Peoples at preventing tropical deforestation? What about other local communities?  In honor of the International Day of the World’s Indigenous Peoples this August 9th, let’s look at some numbers.

The topic of indigenous peoples, local communities, and deforestation received heightened attention recently with the release of a report by the World Resources Institute and the Rights and Resources Initiative entitled, “Securing Rights, Combating Climate Change: How Strengthening Community Forest Rights Mitigates Climate Change.”  The report reviews a large literature and conducts an original analysis of deforestation rates inside and outside of indigenous or community-designated areas in fourteen countries.  It concludes that strengthening the land and resource rights of indigenous peoples and other local communities will reduce deforestation and combat climate change.  The findings have been reported widely, including in Deutsche Welle, The Guardian, Huffington Post, Mongabay, New Scientist, and Scientific American.

There’s good reason to want these conclusions to be true. If strengthening legal recognition of indigenous peoples’ or local communities’ rights to land really does lead to less deforestation, then this represents a simultaneous win for people and a win for forests, and a convenient alignment of social and environmental agendas.  As my colleague Frances Seymour discusses here, the indigenous rights agenda is important on its own terms.  And halting tropical deforestation is critical for maintaining a safe and stable climate.

It's worth examining these arguments with a critical eye, however, because in the past some other conveniently agenda-aligning hypotheses have not held up well to scrutiny. I’m thinking here about claims like “agricultural intensification is sufficient to spare forest from conversion to cropland” (challenged here) and “once people’s incomes rise they will become less interested in clearing forest” (challenged here).

Kalifi Ferretti-Gallon and I recently looked at the connection between indigenous peoples, local communities, and deforestation, as part of a much broader meta-analysis of what drives deforestation and what stops it. We analyzed the relationship between about 40 different variables and deforestation, by aggregating the results from all 117 spatially explicit econometric studies of drivers of deforestation published in peer-reviewed academic journals from 1996-2013.  Unlike a literature review, a meta-analysis systematically takes into account all the available evidence on a particular topic.  And unlike comparisons of deforestation rates inside vs. outside of particular areas, spatially explicit econometric studies are able to control for confounding factors like remoteness or suitability for agriculture that can account for the difference in deforestation rate between areas.  Meta-analyses have their weaknesses too—they are fundamentally limited by the quality and geographic scope of their component studies.  Ours is no exception; more than half of the 117 studies in our database were conducted in just six countries. For more about our study see our CGD Working Paper, policy brief, Wonkcast, and open-access database.

So just how good are indigenous peoples at preventing tropical deforestation? Our database includes variables[1] related to both indigenous territories (which have legally recognized boundaries and rights) and indigenous populations (where this may not always be the case).  On the whole indigenous peoples’ forests were associated with slower deforestation three times as often as they were associated with faster deforestation—a statistically significant difference (Figure 1).  On this point our findings agree with those of the WRI/RRI report: indigenous peoples’ forests are associated with less deforestation.


Figure 1. What Drives Deforestation and What Stops It? A Meta-Analysis (Source: Ferretti-Gallon and Busch, 2014)

Given that indigenous peoples’ forests incur significantly lower deforestation both where territorial rights are legally recognized as well as where they are not, can we necessarily conclude that stronger legal recognition of indigenous rights leads to lower deforestation?  There’s every reason to believe that this is so, and yet we haven’t come across any econometric studies that directly prove this.  Only one study in our database actually had the time-series data on deforestation necessary to analyze changes in deforestation rates within sites before and after official recognition of indigenous peoples’ territorial rights.  This particular study found no significant change, probably because deforestation in such areas was very low both before and after legal recognition.  This study was from Brazil, where the national space agency has been a pioneer in making annual data on the spatial distribution of deforestation publicly available. Now that WRI’s Global Forest Watch has made similar data from Matt Hansen and colleagues available for the entire world, it is possible to conduct studies everywhere on the effect of recognition of indigenous territorial rights on deforestation.  Proving that stronger legal recognition of indigenous land rights forestalls deforestation, either through time series studies or matching studies, should be a priority for future research.

Now, let’s turn to local communities. 

The WRI/RRI report doesn’t actually define what it means by “local communities,” which is problematic since in the absence of context this term could be taken to mean almost anything, from a hunter-gatherer village to a logging camp.  In-the-know readers might intuit that the authors of the WRI/RRI report have used “local communities” to refer to forest communities with a historical and cultural connection to the land and a tradition of sustainable resource use. But even the two organizations involved in the report seem to have different understandings.  In the video of the report launch event, RRI Coordinator Andy White defines “local communities” at 7:35 as, essentially, indigenous peoples in countries where the term “indigenous” has no recognized legal status.  But WRI President Andrew Steer refers to “community forests” in his Yahoo! News op-ed and his speech starting at 11:15 without mentioning indigenous peoples.

Because no definition of “local communities” is made explicit, news articles on the report have ranged widely in their interpretation of the report’s findings.  Several news outlets grasp that the WRI/RRI report authors’ findings relate primarily to indigenous communities (e.g. Scientific American).  But other articles neglect to mention indigenous peoples, referring instead to community-based management, strengthened land tenure (The Guardian), or “letting local people into the forests with the legal right to control what happens there” (New Scientist).

Let’s examine these various interpretations of local communities in turn, starting with community-managed forests. In our meta-analysis, we used the term “community forestry” to refer to variables[2] related to cooperative forest management, or other common property or collective management arrangements, such as ejidos in Mexico.  We found that on the whole community forestry areas were slightly more likely to be associated with lower deforestation than with higher deforestation, but this difference was statistically indistinguishable from what would be expected by repeatedly flipping a coin (Figure 1). 

We found a similar result when we looked at “land tenure security,” which encompassed variables[3] related to more legally secure land ownership rights and duration of occupancy, either for communities or individuals.  Forests with more secure tenure were slightly more likely to be associated with lower deforestation than with higher deforestation, but again, this difference was statistically indistinguishable from a coin flip.

What about multiple-use protected areas? We found that protected areas[4] were associated with lower deforestation six [FN] times as often as they were associated with higher deforestation (as you’d expect, this result is statistically significant). This category includes “strict” protected areas such as national parks that prohibit most forms of exploitation, as well as “multiple-use” reserves that permit some economic activities but not others.  Both types of protected areas are associated with less deforestation far more frequently than more deforestation, by lopsided margins.[5] An active literature exists on which type of protected area is more effective. This debate is clouded by the fact that strict protected areas are generally designated in less accessible, less productive locations than multiple-use protected areas, and thus have lower baseline threat of deforestation. Studies like this one that compare the two types of protection only within areas of similar baseline threat are rare.

Are studies of community forest management only finding what they are looking for? It’s probably worth mentioning that our meta-analysis showed a marked divergence in findings between studies where community forestry was the variable of interest vs. studies where it was included only as a control variable.  In studies where community forestry was the explicit variable of interest, community forestry was associated with lower deforestation 15 times and associated with higher deforestation only once.  But, in studies where community forestry was merely included as a control variable, community forestry was associated with lower deforestation 22 times and associated with higher deforestation 28 times.  We found a similar divergence in studies of land tenure. This suggests that the literature on these topics might be affected by selection bias, in which places with more successful outcomes are more likely to have been studied.

So does this mean that community forest management doesn't slow deforestation? Not at all. It means that community forest management is associated with slower deforestation in some cases (as found by studies from here and here, from Mexico), but not in others (like studies from here and here, also from Mexico). A productive research agenda would be to figure out why community forest management works well in the places where it does, and whether these factors can be replicated elsewhere, rather than taking as dogma that community forest management will always slow deforestation.

Rich countries should pay for forest protection. Indigenous peoples’ management and protection of forest keeps millions of tons of carbon dioxide in trees and out of the atmosphere.  Rich countries have recognized this service through international agreement on a results-based climate mechanism called REDD+, though finance for the mechanism has been slow in coming.  Encouragingly, the $1B pay-for-performance partnership between Norway and Indonesia is making indigenous peoples more visible and recognized than ever before.  As Mina Setra, Deputy Secretary General of Indonesia’s Indigenous Peoples’ Alliance of the Archipelago (AMAN), describes in a CGD Wonkcast.

For indigenous peoples, forest is everything. Many indigenous communities think of the forest as their mother, their blood, their livelihood, the source of everything, the life of the people … for ages, indigenous peoples have been protecting the forest … so actually, REDD+ is not a new thing for indigenous peoples. (2:00)

When REDD+ started, it created controversy for indigenous peoples … but for us, we see REDD+ as an opportunity for us to be seen. You know, ten years ago, no one talked about indigenous peoples, or about their rights. We were trying hard to bring up issues of indigenous peoples’ rights at the national level. It didn’t really work. We tried, in different ways. But when the international community started talking about forests, and about REDD+, then we had the opportunity to say, we do exist! Come on, people, when you talk about forests, you can not escape talking about us ... Nothing about us, without us. (7:20)

It’s also the responsibility of other countries, rich countries, developed countries, to support us and our efforts ... and reduce their own emissions in their own countries. (16:10)

Which variables are associated with lower deforestation?

Indigenous territories: YES
Indigenous populations without recognized territorial rights: YES
Recognition of indigenous territorial rights: Presumably, but few direct econometric studies to date
Community forest management: Sometimes
Strengthened land tenure: Sometimes
Strict protected areas: YES
Multiple-use protected areas: YES
Strict vs. multiple-use protected areas: Unclear

In sum, there really does seem to be something special about the way Indigenous Peoples maintain their forests that doesn’t necessarily transfer to other communities or individuals just because they are “local.”

I’ll discuss these and other findings of our meta-analysis in a webinar hosted by the World Wildlife Fund on Thursday, August 28 at 12pm ET. Register here.


[1] Our evidence base on indigenous peoples comes from 94 regression outputs from 12 studies from Bolivia, Brazil, and Mexico

[2] Our evidence base on community forestry comes from 132 regression outputs from 17 studies from Mexico, El Salvador, Ethiopia, and Guatemala.

[3] Our evidence base on land tenure security comes from 74 regression outputs from 14 studies from 9 countries on 3 continents.

[4] Our evidence base on protected areas comes from 228 regression outputs from 35 studies from 19 countries on 4 continents.

[5] Across all regression outputs, protected areas identified as “strict” are associated with lower deforestation 50 times and higher deforestation 2 times. Protected areas identified as “multiple use” are associated with lower deforestation 42 times and higher deforestation zero times.  Variables associated with other aspects of protection besides strict or multiple use are associated with lower deforestation 75 times and higher deforestation 27 times.

[FN] An earlier version of this blog post stated an incorrect ratio.

Disclaimer

CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.